Atmospheric Research 245 (2020) 105037

Contents lists available at ScienceDirect

Atmospheric Research
journal homepage: www.elsevier.com/locate/atmosres

Long-distance mobile MAX-DOAS observations of NO2 and SO2 over the T

North China Plain and identification of regional transport and power plant
emissions
Wei Tana, Cheng Liua,b,c,h,i,⁎, Shanshan Wangd,e,⁎⁎, Haoran Liuf, Yizhi Zhua, Wenjing Sug,
Qihou Hua, Jianguo Liua
a
Key Laboratory of Environmental Optics and Technology, Anhui Institute of Optics and Fine Mechanics, Hefei Institutes of Physical Science, Chinese Academy of Sciences,
Hefei 230031, China
b
Center for Excellence in Regional Atmospheric Environment, Institute of Urban Environment, Chinese Academy of Sciences, Xiamen 361021, China
c
Department of Precision Machinery and Precision Instrumentation, University of Science and Technology of China, Hefei 230026, China
d
Shanghai Key Laboratory of Atmospheric Particle Pollution and Prevention (LAP3), Department of Environmental Science and Engineering, Fudan University, Shanghai
200433, China
e
Institute of Eco-Chongming (IEC), No.20 Cuiniao Road, Shanghai 202162, China
f
Institute of Physical Science and Information Technology, Anhui University, Hefei 230601, China
g
School of Earth and Space Sciences, University of Science and Technology of China, Hefei 230026, China
h
Key Laboratory of Precision Scientific Instrumentation of Anhui Higher Education Institutes, University of Science and Technology of China, Hefei 230026, China
i
Anhui Province Key Laboratory of Polar Environment and Global Change, USTC, Hefei 230026, China

ABSTRACT

In this study, the spatiotemporal distribution of the NO2 and SO2 Vertical Columns Densities (VCDs) in the North China Plain (NCP) region was measured by long-
distance mobile measurements using a mobile Multi-AXis Differential Optical Absorption Spectroscopy (MAX-DOAS) instrument. The mobile observations were
performed in both summer (July 2017) and winter (January and February 2018) and the total driving mileage exceeded 3000 km. The concentrations of NO2 and SO2
in different seasons and areas were significantly different. During winter, serious NO2 and SO2 pollution was observed in northern Anhui Province, central Shandong
Province, and the Beijing-Tianjin-Hebei Region. The evolution and transportation processes of three typical heavy pollution cases were discussed in detail. Combined
with the Weather Research and Forecasting with Chemistry model simulated wind field information, the NO2 transportation flux from northern Jiangsu Province to
northern Anhui Province is quantified as 7.12 kg s−1. Finally, we estimated the NO2 and SO2 emissions from the Dezhou and Hengshui power plants by plume cross
section scanning and encircled observations, respectively. The NO2 and SO2 emission fluxes of the Dezhou Power Plant are 0.79 kg s−1 and 1.11 kg s−1, respectively,
while the NO2 and SO2 emission fluxes of the Hengshui Power Plant are 0.12 kg s−1 and 0.36 kg s−1, respectively. This study quantitatively analyzed the trans-
portation of atmospheric pollutants and power plant emissions, which is helpful for understanding the occurrence and evolution of pollution and is also useful for the
government to introduce policies to protect and control the atmospheric environment.

1. Introduction
The North China Plain (NCP) region is bounded by the Yan
Mountains (north), Dabie Mountains and Yangtze River (south), Taihang
Mountains (west), and Bohai and Huanghai seas (east). This region
contains the major parts of Hebei, Henan, and Shandong Provinces and
the northern parts of Anhui and Jiangsu Provinces. The megacities of
Beijing and Tianjin are also included in this region. Although the land
area of NCP region only takes up 3.22% of the total China land area, but
the area of cultivated land in this region accounts for about 20% of the
total area of cultivated land in China. Furthermore, about 20% of the
total population in China lives in this region. Owing to rapid in-
dustrialization and urbanization, the NCP region is facing severe atmo-
spheric pollution problems, especially the Beijing-Tianjin-Hebei region. It
usually suffers from severe haze pollution combined with high con-
centrations NO2 and SO2 pollution events during winter. Moreover, the
heavy regional pollution has spread to an even larger area, lasted for a
long time, and seriously impacted the lifestyle and health of residents
(Zhao et al., 2013; Fu et al., 2014; Sun et al., 2014; Wu et al., 2017a; Su,
2018; Zhang et al., 2018; Wang et al., 2019).

⁎
Correspondence to: C. Liu, Department of Precision Machinery and Precision Instrumentation, University of Science and Technology of China, Hefei 230026,
China.
⁎⁎
Correspondence to: S. Wang, Shanghai Key Laboratory of Atmospheric Particle Pollution and Prevention (LAP3), Department of Environmental Science and
Engineering, Fudan University, Shanghai 200433, China.
E-mail addresses: chliu81@ustc.edu.cn (C. Liu), shanshanwang@fudan.edu.cn (S. Wang).

https://doi.org/10.1016/j.atmosres.2020.105037
Received 24 December 2019; Received in revised form 13 April 2020
Available online 16 May 2020
0169-8095/ © 2020 Published by Elsevier B.V.
W. Tan, et al.
Table 1
The detail information of mobile measurement operations.
Date Colour of route Observation route
August 2017 21 Blue, in Fig. 1 (a) BJ-TJ-CZ-DZ-JN-TA-JNI-ZZ- XZ-(HB-SZ-BB)
22 Red, in Fig. 1 (a) BB-HN-HF
January 2018 15 Blue, in Fig. 1 (b) HF-HN-BB-SZ-HB-(XZ-ZZ)
16 Green, in Fig. 1 (b) ZZ-JNI-TA-JN-DZ-CZ-LF-BJ
17 Red, in Fig. 1 (b) BJ-BD-HS
24 Green, in Fig. 1 (c) HS-DZ-JN
27 Green, in Fig. 1 (c) JN-DZ-HS
February 2018 09 Green & Blue, in Fig. 1 (c) HS-DZ-JN-TA-JNI-(ZZ-XZ-HB-SZ)
10 Red, in Fig. 1 (c) SZ-BB-HN-HF
* The observations under the solar zenith angle (SZA) larger than 80° are marked in brackets, and the different cities are represented by abbreviated capital letters
(BJ: Beijing, TJ: Tianjin, LF: Langfang, CZ: Cangzhou, BD: Baoding, HS: Hengshui, DZ: Deizhou, JN: Jinan, TA: Taian, JNI: Jining, ZZ: Zaozhuang, XZ: Xuzhou, HB:
Huaibei, SZ: Suzhou, BB: Bengbu, HN: Huainan, HF: Hefei).
The sources emissions, regional pollutant transport, chemical pro-
cesses, and unfavorable meteorological conditions for diffusion have
been identified as the most important causes of the heavy pollution
events (Chan, 2017). Owing to rapid development and dense anthro-
pogenic activities, there are many industrial, power, transport, and
residential emissions in the NCP region (Xing et al., 2011; Zhang et al.,
2018; Chen et al., 2019). Therefore, the government has recently for-
mulated environmental protection policies, such as adjusting the in-
dustrial structure, shutting down polluting factories, restricting vehicle
use through license plate rules during heavy pollution periods, and
other integrated regional management measures (MEP, 2017). Mean-
while, the regional transportation of pollutants is significantly influ-
enced by meteorological conditions such as the wind speed, wind di-
rection, atmospheric temperature and pressure conditions. The static
and stable meteorological conditions can easily cause the accumulation
of atmospheric pollutants, and are not conducive to the diffusion and
transportation of pollutants. However, the atmospheric pollutants can
be also transported by horizontal advection, which may cause atmo-
spheric pollution over larger areas (Xu et al., 2011; Jin et al., 2016; Wu
et al., 2018). Moreover, atmospheric chemical processes are also a
major reason for secondary aerosol, O3, and some other pollution. With
the favorable atmosphere conditions for chemical reactions, the im-
portant precursors of NO2 and SO2 tend to form nitrate and sulfate, and
finally form secondary aerosols (Jang and Kamens, 2001; Huang et al.,
2014).
With the advantages of spatial and temporal coverage, satellite
observations are widely used to investigate large-scale pollution pro-
cesses and long-distance pollutant transport cases (Tao et al., 2012;
Wang et al., 2012; Zhang et al., 2018; Zhang et al., 2019). However, this
method does not have sufficient sensitivity to the surface layer (lower
troposphere), where the main atmospheric pollutants are located. As a
detection method with high sensitivity and continuity, Multi-AXis Dif-
ferential Optical Absorption Spectroscopy (MAX-DOAS) technique has
been widely employed in recent years. As one of the applications of the
DOAS method (Platt and Stutz, 2008), it is widely used to retrieve the
vertical distribution and total columns of trace gases, e.g., NO2, SO2,
HCHO, HONO, and aerosols (Clémer et al., 2010; Irie et al., 2011;
Pinardi et al., 2013; Wang et al., 2014; Chan et al., 2015; Chan et al.,
2017; Chan et al., 2019). In addition, some MAX-DOAS studies have
also evaluated trace gas horizontal distribution in recent years (Ortega
et al., 2015; Casaballe et al., 2017; Seyler et al., 2019; Casaballe et al.,
2020). Owing to its portability, the MAX-DOAS instrument can be
carried on mobile platforms to obtain the spatiotemporal distribution of
atmospheric pollutants over large areas and obtain the emission char-
acteristics by encircling or scanning the pollution sources (Ibrahim
et al., 2010; Shaiganfar et al., 2011; Shaiganfar et al., 2017; Tan et al.,
2019).
In this study, a MAX-DOAS instrument was on board a passenger
car, and used to measure the tropospheric VCD of NO2 and SO2 over
NCP region. Based on the mobile MAX-DOAS observations, we studied
2
Atmospheric Research 245 (2020) 105037
Observation time (LT) Observation distance (km)
08:04–17:57 712
08:21–10:43 159
13:18–16:25 283
08:17–16:10 642
09:12–15:18 291
14:05–16:26 207
13:17–16:14 225
13:16–16:50 305
07:56–11:50 243
the pollution characteristics and related transportation processes in the
NCP region in summer (August 2017) and winter (January and
February 2018). The long-distance mobile MAX-DOAS observations
revealed the spatial distribution characteristics of the NO2 and SO2
tropospheric vertical column densities (VCDs) over large areas of the
NCP region. In addition, we also quantitatively analyzed the typical
pollutant transportation from northern Jiangsu Province to northern
Anhui Province during the pollution cases. Finally, the emissions of NO2
and SO2 from the typical point sources of the Hengshui and Dezhou
power plants were estimated.
2. Methodology
2.1. Mobile MAX-DOAS measurements
2.1.1. Routes of mobile MAX-DOAS observations
In this campaign, mobile observations were conducted over a total
of 9 days, including 2 days in summer (August 2017) and 7 days in
winter (January and February 2018). The details of the daily mea-
surement routes and distances are listed in Table 1. As shown in Fig. 1,
the mobile measurements crossed large parts of the NCP region, and the
total driving mileage exceeded 3000 km. The distance from Hefei at the
southernmost end to Beijing at the northernmost end is about 1000 km.
These routes pass through more than 17 cities. Fig. 1(a), (b), and (c)
present the three completed long-distance north to south observations,
which were implemented on 21 to 22 August 2017, 15 to 17 January
and 09 to 10 February 2018, respectively.
2.1.2. Experimental setup
During the summer and winter observations, two different designed
MAX-DOAS instruments were used. For the summer observations a
MAX-DOAS instrument combined with a GPS recorder were applied on
a passenger car, as described in (Tan et al., 2019). A compact Airyx
MAX-DOAS instrument which was special design for mobile observa-
tion has been used in winter observations. The compact MAX-DOAS
system has integrated a GPS recorder, a laptop computer and an MAX-
DOAS instrument. This system has been used in an ship-based study in
the East China Sea (Tan et al., 2018), and similar instruments have also
been used in relevant studies (Kreher et al., 2019; Wang et al., 2020).
Both the two MAX-DOAS instruments consist of a spectrometer, a
temperature controlling unit, a scanning telescope, a connecting fiber,
controlling electronic devices, and other necessary devices. The spec-
trometers of both them are Avantes spectrometer (AvaSpec-ULS2048L-
USB2) cover the same wavelength range of 300 nm to 460 nm with a
Full Width Half Maximum (FWHM) spectral resolution of 0.6 nm, with
a one-dimensional CCD detector (Sony ILX511, 2048 individual pixels).
However, compared with the former instrument the latter one opti-
mized the performance of some configurations, such as the field of view
(FOV) of telescope changed from 0.5 to 0.3 and the type of optical fiber
changed from single-core fiber to multi-core optical fiber. The signal-to-
W. Tan, et al.
Fig. 1. The routes of mobile MAX-DOAS measurements, which are indicated by different colors for the observations during different days. (a) shows the summer
observations in August 2017, while (b) and (c) are the winter observations in January and February 2018.
noise ratio (SNR) of the latter instrument is better than that of the
former, especially in short wave radiation. For the summer measure-
ments in August 2017, the measurement sequence of elevation angles
(α) of 1 × 90° and 4 × 30° was chosen. With the temporal resolution of
30 s for individual spectrum measurement, a full scanning sequence
took about 3 min in total. During the winter observations from 15 to 17
January 2018, the elevation angle scanning sequence was set to be
1 × 90° and 3 × 30°. An individual spectrum measurement took about
20 s, and the whole scanning sequence duration about 1.5 min. To
achieve the higher temporal resolution of zenith measurements, the
elevation scanning sequence was changed to the combination of
1 × 90° and 1 × 30° from 24 January 2018. The duration of an in-
dividual spectrum measurement was about 30 s, while each scanning
sequence took about 1 min. All of these elevation angle settings include
zenith (90°) and low elevation angle (30°) observations. At first the
mobile observations were performed along the expressway, which far
away from some larger emission sources, the zenith observations were
set up in low frequency. However, the frequency of zenith measure-
ments has increased during the observations close to important emis-
sion sources such as cities and power plants. The daily measured scat-
tered sunlight spectra were automatically collected in a laptop
computer. In addition, a high-precision Global Position System (GPS)
data receiver was used to record the real-time coordinate positions
information of the mobile measurements.
2.1.3. DOAS retrieval
The measured spectra were analyzed through the QDOAS spectral
fitting software suite, which is based on the DOAS technique and de-
veloped by BIRA-IASB (http://uv-vis.aeronomie.be/software/QDOAS/
). Before the DOAS fitting, the measured spectra were first corrected by
the offset and dark current. Furthermore, the zenith measurements of
each sequence were used as the Fraunhofer reference spectrum. The
details of the retrieval settings are listed in Table 2. The wavelength
ranges of 338 nm to 370 nm and 312 nm to 326 nm were chosen for
NO2 and SO2 retrieval, respectively. For the analysis of NO2, the trace
gas absorption cross sections of O4 at 293 K (Thalman and Volkamer,
2013), NO2 at 220 K and 298 K (Vandaele et al., 1998), HCHO at 297 K
3
Atmospheric Research 245 (2020) 105037
(Meller and Moortgat, 2000), O3 at 223 K and 243 K (Serdyuchenko
et al., 2014), and BrO at 223 K (Fleischmann and Hartmann, 2004)
combined with a Ring spectrum were included in the DOAS fitting
process. Different to the retrieval of NO2, we included the trace gas
absorption cross section of SO2 at 298 K (Vandaele et al., 2009) and
removed the cross sections of O4 at 293 K and NO2 at 220 K for the SO2
analysis. A 5-order polynomial was included in the DOAS fitting for the
two trace gases. Moreover, the wavelength calibration was achieved by
fitting the measured zenith spectrum to a high resolution solar spec-
trum (Chance and Kurucz, 2010). These NO2 retrieval settings have
been used during Second Cabauw Intercomparison campaign for Ni-
trogen Dioxide measuring Instruments (CINDI-2) campaign (Kreher
et al., 2019), while the SO2 retrieval settings also referred to the pre-
vious relevant study (Theys et al., 2015).
As the measured zenith spectrum of each elevation angle sequence
was selected as the Fraunhofer reference spectrum, the DOAS retrieval
yielded the Differential Slant Column Densities (DSCDs). A typical
DOAS spectral fitting process of the spectrum measured at the elevation
of 30° at 13:59 local time (LT) on 15 January 2018 is presented in
Fig. 2. The retrieved DSCDs of NO2 and SO2 are 3.18 × 1016 molec
cm−2 and 3.52 × 1016 molec cm−2, respectively. The clear absorption
structure and relatively low residuals indicate the good performance of
spectral fitting. Owing to the influenced by sunlight blocked by build-
ings and other objects, there are some failure fitting results. For the NO2
DSCDs under SZA < 80° and the DOAS fit root mean square
(RMS) < 0.003 were retained, while the SO2 DSCDs under SZA < 70°
and the RMS < 0.005 were included in the discussion. Afterwards,
there are about 88% and 64% DSCD results remains for NO2 and SO2,
respectively.
2.1.4. Retrieval vertical column densities of trace gases
To obtain the VCD results, the DSCDs were converted via the Air
Mass Factors (AMFs) using the following formula:
SCDtrop DSCD
VCDtrop = =
AMFtrop AMF 90
°
AMF = 90
° (1)
where α is the elevation angle of the measurement. Besides the zenith
W. Tan, et al. Atmospheric Research 245 (2020) 105037

Table 2
DOAS spectral fitting settings of NO2 and SO2.
Parameter Data source Trace gases

NO2 SO2

Wavelength range 338–370 nm 312–326 nm

O4 Thalman and Volkamer (2013), 293 K √ ×
NO2 Vandaele et al. (1998), 220 K, 298 K, I0-correction (1017 molec cm−2) √ √ (only 298 K)
SO2 Vandaele et al. (2009), 298 K × √
HCHO Meller and Moortgat (2000), 297 K √ √
O3 Serdyuchenko et al. (2014), 223 K, 243 K, I0-correction (1020 molec cm−2) √ √
BrO Fleischmann et al. (2004), 223 K √ √
Ring Ring spectra calculated with QDOAS according to Chance and Spurr (1997) √ √
Polynomial degree 5 5
Wavelength calibration Based on a high resolution solar reference spectrum (SAO 2010 solar spectra)

observations, there were only 30° measurements during the mobile
campaign. Thus, the AMF can be calculated easily and quickly through
the geometric approximation method, which is a widely used method
that has been validated by the radiative transfer model (RTM) calcu-
lation (Hönninger et al., 2004; Brinksma et al., 2008; Wagner et al.,
2010; Halla et al., 2011; Tan et al., 2018).
AMF = 1 sin( )
As calculated by Eq. (2), the AMF of the 30° elevation angle is 2
while the AMF of the 90° elevation angle is 1; thus, the tropospheric
VCDs of the trace gases are approximately equal to the DSCDs for the
measurements under the elevation angle of 30°.
et al., 2018). In addition, this product has been validated to be more
representative over China because more accurate NO2 a priori profiles
were used in the AMF calculation (Liu et al., 2016; Su et al., 2017; Xing
et al., 2017; Hong et al., 2018; Tan et al., 2018).
2.3. Estimation of pollutant transportation and source emissions
(2)
In this study, mobile observations were conducted to estimate the
pollutant transportation and source emissions over the NCP region.
Combined with the wind information, the pollutant transportation and
source emission fluxes can be acquired through the integral along the
driving route using the following equation (Ibrahim et al., 2010):

2.2. Ozone Monitoring Instrument observations Flux = VCD (s ) n ds (3)

The Ozone Monitoring Instrument (OMI) was onboard the NASA
Earth Observing System Aura satellite in 2004 (Levelt et al., 2006). The
OMI is an imaging spectrometer that covers the spectral range from
270 nm to 500 nm. The OMI measured spectra are used to retrieve data
on O3, NO2, and other trace gases. The overpass time of the satellite is
about 13:45 LT. The USTC's OMI tropospheric NO2 product was com-
pared with the mobile measurements. It is developed based on the
OMI's primary product through the DOAS fitting of OMI spectra (OMI
Level 1B VIS Global Radiances Data Product; https://disc.gsfc.nasa.
gov/Aura/dataholdings/OMI/oml1brvg_v003.shtml). The Slant
Column Densities (SCDs) of NO2 are separated to stratospheric and
tropospheric columns based on the local analysis of the stratospheric
field over unpolluted areas (Bucsela et al., 2013; Krotkov et al., 2017).
For the conversion of tropospheric OMI NO2 SCDs to VCDs, we re-
calculated the AMF based on the NO2 and atmospheric profiles derived
from the Weather Research and Forecasting with Chemistry (WRF-
Chem) model simulations (Grell et al., 2005; Gao et al., 2016; Zhou
where VCD indicates the tropospheric VCD of trace gases, is the wind
vector, and n is the unit vector orthogonal to the driving route. For the
measurements conducted in segments, the transportation and emission
flux of trace gases can be obtained by the following equation:
Flux = VCDi vi sin i si (4)
where VCDi is the VCD of trace gases of segment i, vi is the wind speed
corresponding to segment i, θi is the angle between the wind direction
and driving direction of segment i, and ∆si is the distance between the
location of the two succeeding measurements i and i + 1.
The wind field information was obtained through the WRF-Chem
simulation. First, the simulated high spatial (1 km × 1 km) and tem-
poral (10 min) resolution 3-dimensional wind field data were weight-
averaged by the WRF-Chem model-simulated trace gas profiles at dif-
ferent vertical heights (0–2 km). The horizontal wind field information
was then interpolated to the position of the mobile measurements.

Fig. 2. Typical DOAS spectral fittings for (a) NO2 and (b) SO2. The spectrum was collected at an elevation of 30° at 13:59 LT on 15 January 2018. The red curves
show the reference absorption cross section scaled to the measured atmospheric spectrum (black curves) by DOAS fitting.

4
W. Tan, et al.
Fig. 3. The spatial distributions of NO2 and SO2 VCDs were measured on (a) 21–22 August 2017, (b) 15–17 January 2018 and (c) 09–10 February 2018 over NCP
region, China. The driving routes are marked by the dark thin lines, while the trace gases VCDs are indicated by colored thick lines.
3. Results and discussion
3.1. Spatial distribution of tropospheric NO2 and SO2 VCDs
Combined with the GPS logged position information, we obtained
the VCDs of different trace gases along the observation routes. Fig. 3
shows the spatial distributions of the NO2 and SO2 VCDs of the three
continuous long-distance observations crossing from the edge of the
Yangtze River Delta region to the NCP region. In addition, the round-
trip measurements in the center of the NCP region from Hengshui to
Jinan passing through Dezhou are presented in Fig. 4. The dis-
continuous VCD plots are caused by the filter of the high SZA and RMS
results.
It is obvious that the NO2 pollution in winter was more severe than
that in summer during this study. The NO2 VCDs vary from 3.54 × 1015
molec cm−2 to 93.50 × 1015 molec cm−2 with a mean value of
35.97 × 1015 molec cm−2 in winter, while they range from
0.22 × 1015 molec cm−2 to 49.60 × 1015 molec cm−2 in summer with
an average of 11.41 × 1015 molec cm−2. Higher levels of NO2 (~3
times) was obtained in winter, which can be explained by the more
fossil fuel combustion for heating and the bad conditions for the spread
5
Atmospheric Research 245 (2020) 105037
of pollutants in winter (Ma et al., 2013; Hendrick et al., 2014). The SNR
of the instrument used in summer observations is relatively low in short
wave radiation where is the strong absorption wavelength range of SO2.
Meanwhile, the SO2 concentrations are low in summer. Therefore, the
retrieved SO2 DSCDs were removed during the filtering of DOAS fit
RMS and DSCD errors. The SO2 VCDs vary from 0.11 × 1015 molec
cm−2 to 105.00 × 1015 molec cm−2 with an average of 29.47 × 1015
molec cm−2 in winter. These results were comparable to the previous
long-term ground-based MAX-DOAS measurements in this region, in
which high values around 40–70 × 1015 molec cm−2 were observed in
winter while the minimum of which around 20 × 1015 molec cm−2
appeared in summer (Wang et al., 2014).
In the upper panel of Fig. 3 and Fig. 4, the spatial distribution of
NO2 VCDs shows the three high NO2 pollution regions located in
northern Anhui Province, central Shandong Province, and the route
from Beijing passing by Baoding and ending in Hengshui. Furthermore,
the highest NO2 VCDs (i.e., > 81.30 × 1015 molec cm−2) were ob-
served on the way from Beijing to Hengshui on 17 January 2017. In
general, this is the heavily polluted area in the Beijing-Tianjin-Hebei
region of China, and the detailed explanation of this pollution process is
provided in Section 3.3. In addition, the high value of NO2 VCDs were
W. Tan, et al.
Fig. 4. The spatial distributions of NO2 and SO2 VCDs were measured on (a) 24 January 2018 and (b) 27 January 2018 over center NCP region, China. The driving
routes are marked by the dark thin lines, while the trace gases VCDs are indicated by colored thick lines.
obtained usually when the car driving closer to the city center. The
spatial distribution of SO2 VCDs during the campaign is reported in the
bottom panel of Fig. 3 and Fig. 4, in which the three major heavily
polluted regions are the same as the hotspots of NO2 distribution.
However, the highest SO2 VCDs (i.e., > 88.20 × 1015 molec cm−2)
were measured in Shandong Province centered at Jinan, Taian, and
Jining on 16 January 2018. As there are many power plants in these
areas, these high SO2 VCDs were expected (Xing et al., 2011).
Furthermore, we compared the mobile observations and OMI sa-
tellite products to validate the spatial distribution of NO2 VCDs. First,
the satellite data were filtered with the criteria of a relative error of less
than 100% and the could fraction of less than 0.4. The daily spatial
distributions of NO2 VCDs observed by the OMI and mobile MAX-DOAS
were then plotted together in Fig. 5 for the days of 21 August 2017, 17,
24 January and 09 February 2018. It is obvious that the spatial dis-
tributions of the two datasets agreed well with each other, especially for
the OMI overpass time. However, the mobile observations also show
higher sensitivity to the surface emission sources, such as large cities,
power plants, and refineries. Moreover, the pollution of large areas was
reported by both the OMI and mobile measurements on 17, 24 January
and 09 February 2018.
For a better match in temporal and spatial coverage, the mobile
observed NO2 VCDs close to the OMI overpass time (12,00–14:00 LT)
were averaged by the grid of 0.1° × 0.1°, while the OMI results were
averaged at the same grid. Fig. 6 shows the correlation analysis of these
4 days' observations in Fig. 5. Besides the correlation analysis between
6
Atmospheric Research 245 (2020) 105037
the mobile MAX-DOAS NO2 VCDs and USTC's OMI NO2 product, the
mobile MAX-DOAS measured NO2 VCDs have been also compared with
the NASA's standard OMI NO2 product for reference. The Pearson cor-
relation coefficient (R) between the mobile measurement and USTC
product is 0.92 which is slightly higher than the Pearson correlation
coefficient of 0.91 for the comparison with the NASA's standard pro-
duct. In addition, the USTC product underestimated the tropospheric
NO2 VCD by about 25% on average while this for NASA's standard
product increased to 40%. In general, the USTC product which uses the
NO2 profiles calculated from WRF-Chem simulations as a priori profile
shows a better agreement with the mobile measurements. However, for
the mobile observations, which are more sensitive to pollution sources,
there are wide ranges and variations in the mobile measurements while
the satellite products with a slight variation, such as the observation on
21 August 2017. The mobile measurements range from 5.05 × 1015
molec cm−2 to 33.40 × 1015 molec cm−2 with a mean of
(10.06 ± 6.52) × 1015 molec cm−2 while the satellite observations
vary from 5.25 × 1015 molec cm−2 to 9.42 × 1015 molec cm−2 with an
average of (7.69 ± 1.32) × 1015 molec cm−2.
3.2. Observed NO2 and SO2 over different cities
In order to characterize the air pollution in a large area, we com-
pared the NO2 and SO2 VCDs measured by mobile MAX-DOAS with the
surface NO2 and SO2 concentrations measured by in-situ monitors over
different cities of the NCP. The in-situ monitoring data is hourly
W. Tan, et al. Atmospheric Research 245 (2020) 105037

Fig. 5. Comparison of OMI and mobile MAX-DOAS measured NO2 VCDs on (a) 21 August 2017, (b)17 January 2018, (c) 24 January 2018, and (d) 09 February
2018.The mobile measurements were plotted overlap in the base map of OMI products. The routes of mobile measurements are indicated by the white lines, and the
position of the mobile measurements at 12:00 and 14:00 LT are marked by black triangles.

resolution (National Air Quality Historical Data, http://beijingair.
sinaapp.com/data/beijing/all/yyyymmdd/csv). The mobile MAX-
DOAS results were first hourly averaged. The value of the nearest
segment (within 25 km) to the city center was then chosen to compare
with the in-situ observation results, which were the averages of mul-
tiple measurement sites in the cities during the same time period when
the mobile observation car passed by. As shown in Fig. 7, the con-
centrations of NO2 and SO2 in winter are obviously higher than those in
summer, which were detected by both mobile MAX-DOAS and in-situ
observations. The time series of the NO2 levels of the two datasets are
consistent well in Fig. 7(a), with a Pearson correlation coefficient (R) of
0.64 (Fig. 8(a)). Different from the NO2 comparison, much more dif-
ferences of the two datasets are presented for SO2 comparison in
Fig. 7(b), and the Pearson correlation coefficient reduce to 0.42
(Fig. 8(b)). In addition, the reasonable explanation for the deviations
between these two methods for some cases could be that the city center
or the in-situ monitoring sites are far from the mobile observation
routes. As shown in Fig. 7, the in-situ data report both peak values of
NO2 and SO2 in Jinan on 16 January 2018, but the relatively lower
values are obtained from the mobile MAX-DOAS observations. As the
closest mobile measurement in this day is about 21 km (Fig. 7 (c)) far
from the city and the mobile MAX-DOAS observations are insensitivity
to this far distance. In addition, the wind direction is also a strong in-
fluencing factor, for the comparison between mobile MAX-DOAS ob-
servations at the up-wind and down-wind areas of city may have sig-
nificant difference. Meanwhile, here are also some explanations for the
difference in NO2 and SO2 comparison. First, the emission sources of
NO2 and SO2 are quite different. The NO2 pollution is mainly from the
traffic, industrial and power plant emissions. As the traffic emissions
are widely distributed in the city, the in-situ monitor can observe the
NO2 emissions sensitively. However, the SO2 pollution is major come
from industrial and power plant emissions in the city. The SO2 pollution
is usually distributed near the factories and power plants, the spatial
distribution of the SO2 pollution may not be as balanced as that of NO2
pollution. Therefore, the mobile MAX-DOAS observation location far
away from the in-situ monitoring may lead to larger difference in the
trends of SO2 VCDs and surface concentration. Second, the vertical
distribution of NO2 and SO2 concentration is different. Usually, the NO2
pollution is mainly concentrated in the near surface, and the con-
centration decreases with the increase of height. Thus, the variation
trend of VCDs is consistent well with the surface concentration detected
by in-situ monitor. However, the SO2 pollution are original from the

7
W. Tan, et al.
Fig. 6. Correlation analysis of mobile MAX-DOAS and OMI measured NO2
VCDs. The error bars show the OMI error and standard deviation of mobile
MAX-DOAS observations. Both the USTC's (blue markers) and NASA standard
(red markers) OMI product were used in the correlation analysis.
elevated point source discharge, and the high value is not located at
surface (Wang et al., 2014; Smekens et al., 2015). Therefore, the var-
iation trend of VCDs may different from the surface concentration.
The ratios of SO2 to NO2 of the cities passed by during the mobile
observation campaign were presented in Fig. 9. We compared the ratios
of SO2 to NO2 measured by the mobile MAX-DOAS instrument (red line
in Fig. 9) with those from the simultaneous in-situ measurements (black
line in Fig. 9) and also compared these with the winter average (blue
line in Fig. 9) and the yearly average (pink line in Fig. 9) of the in situ
monitor. Except for some special conditions, the ratios of SO2 to NO2
from the two methods show a nearly consistent variation trend.
Moreover, the values obtained from the in-situ monitor are normally
lower than those obtained from mobile observations. Because the SO2
pollution usually originates from elevated point source emissions, the
higher concentrations of SO2 may be located at higher positions, which
Fig. 7. Comparison of (a) NO2 and (b) SO2 results measured by mobile MAX-DOAS and in-situ measurement sites over different cities of NCP areas, China.
Meanwhile, the distances between the mobile MAX-DOAS observation and city center are presented in figure (c).
8
Atmospheric Research 245 (2020) 105037
will result in a relatively lower SO2/NO2 ratio at the surface position.
From the comparison of the winter averaged (mean value of January
and February) and yearly averaged ratios of SO2 to NO2, it is clear that
the in-situ surface SO2/NO2 ratios in winter are higher than the annual
average, which is attributed to the high consumption of coal fuels
during the winter heating season.
From the in-situ observations, lower SO2/NO2 ratios were found in
Beijing, which is the capital of China. As increasing attention has been
focused on environmental protection and sustainable development in
recent years, factories and power plants are constantly exploring new
production and operation modes, such as the replacement of thermal
power plants by the gas power plants in Beijing. However, the NO2
pollution mainly originates from vehicle emissions, which are not easily
controlled, except for by traffic restrictions. These factors led to a lower
SO2/NO2 ratio in Beijing. In addition, high SO2/NO2 ratios have been
repeatedly detected in Shandong Province, as observed in Jinan, Taian,
and Zaozhuang, and high concentrations of NO2 and SO2 pollution is
also shown in these cities in Fig. 7. The high SO2/NO2 ratios are mainly
caused by the more SO2 emissions from the energy consumption pro-
cess. Shandong is a large energy-consuming province with a large
number of thermal power plants and developed industries. According to
the special topographical characteristics of Shandong Province, the
phenomenon of temperature inversion occurs frequently, which
weakens the convective motion of the air and is unfavorable for the
diffusion of atmospheric pollutants (Xu et al., 2011; Tao et al., 2012;
Zhao et al., 2013; Chen et al., 2019). Therefore, the accumulation of air
pollution easily occurred in this area.
3.3. Regional pollutant transport
As described in Section 3.1, the three pollution cases of high con-
centrations of NO2 and SO2 in winter were found in northern Anhui
Province on 15 Jan 2018, central Shandong province on 16 Jan 2018
and the Beijing-Tianjin-Hebei region on 17 Jan 2018. The NO2 VCDs of
these three cases measured by the mobile MAX-DOAS instrument are
shown in the upper panel of Fig. 10(a) to (c). To reveal the evolution
processes of these pollution cases, we used the WRF-Chem model si-
mulation to obtain the wind field and pollutant concentration in-
formation. The simulated hourly averaged surface NO2 concentrations
W. Tan, et al.
Fig. 8. The correlation analysis of mobile MAX-DOAS observations and in-situ measurements over different cities of NCP areas for (a) NO2 and (b) SO2, and the data
are accorded with the comparison in Fig. 7.
and wind field results corresponding to the highest NO2 VCD periods in
Fig. 10(a), (b), and (c) are presented in Fig. 10(d), (e), and (f), re-
spectively.
In the first case shown in Fig. 10(a) and (d), owing to the large
number of thermal power plants and factories located in northern
Jiangsu Province, local emissions are the major reason for the heavy
pollution. Under the easterly wind, the atmospheric pollutants from
northern Jiangsu Province were brought to northern Anhui Province. In
addition, the pollutants were even spread to the areas adjacent to
Henan Province. In the second case (Fig. 10(b) and (e)), the northerly
wind is dominant during the pollution period in central Shandong
Province. Under the action of northerly wind, the atmospheric pollu-
tants in the Beijing-Tianjin-Hebei region may be transported to Shan-
dong Province. Moreover, the low wind speed in central Shandong
Province is unfavorable for the transport and spread of pollutants. In
the last case (Fig. 10(c) and (f)), the shape of the heavily polluted areas
of the Beijing-Tianjin-Hebei region looks like a strip. The northwest
wind from northern Hebei Province and the southwest wind from
Shandong Province simultaneously forced the pollutants to accumulate
Fig. 9. The ratios of SO2 to NO2, which were obtained by in-situ and mobile MAX-DOAS measurements.
9
Atmospheric Research 245 (2020) 105037
here. Under the southeast wind, the pollutants spread in this area.
According to previous studies, this type of pollution process is very
common in this region (Xu et al., 2011; Fu et al., 2014).
To quantify the transport of pollutants, the transportation flux es-
timation method described in Section 2.3 was employed to calculate the
flux. Fig. 11 shows the pollution case measured in northern Anhui
Province on 15 January 2018, where the colored line is the NO2 VCDs
along the mobile observation route. The hourly wind field information
simulated by the WRF-Chem model during 15:00–16:00 LT is indicated
by the black arrows, and the NO2 VCDs between the two black triangles
mark the positions measured during the same period. This shows that
the wind between the cities of Huainan (HN) and Huaibei (HB) is
mainly from the east and almost vertical to the driving route. Therefore,
the segment (from 32.5° N to 34.0° N with a distance of about 200 km,
take about 1.5 h.) included in the black box was chosen for the esti-
mation of the total transportation across the route. The total transpor-
tation flux can be obtained through the integral along the observation
route using Eq. (4) in Section 2.3.
In addition, there is another method to acquire the total
W. Tan, et al. Atmospheric Research 245 (2020) 105037

Fig. 10. The spatiotemporal distributions of NO2 concentrations during the three typical pollution cases, which (a) is the NO2 VCDs measured by mobile MAX-DOAS
on 15 Jan 2018, (b) and (c) is same to (a) but for the observations on 16 and 17 Jan 2018, respectively, while the (d), (e) and (f) are the model simulated hourly
averaged surface NO2 concentration and wind field results corresponding to the highest polluted segment of (a), (b) and (c) (between the two short black lines),
respectively. The wind field information is indicated by black arrows.

transportation flux based on the simulated NO2 profile and wind in-
formation, as described in Eq. (5). The simulated NO2 VCDs at every
height layer are multiplied by the wind speed at the same grid and then
sum up from the surface to the height of 2 km. By obtaining the
transportation of a single observation segment, the total transportation
flux is calculated by summing the transportation of all individual ob-
servations.
Flux = cji vji sin ji hj si
i j
where cji is the concentration of NO2, vji is the wind speed, θji is the
angle between the wind direction and the driving direction, δhj is the
height of layer j, and ∆si is the distance between the two successfully
measured segments.
Consequently, both these two methods were used in the calculation
of the transportation flux of the pollution case in northern Anhui
Province. Fig. 12(a) shows the transportation of every individual ob-
servation segment between HN and HB, which are calculated by the
mobile MAX-DOAS measured NO2 VCD and the model simulated wind
information. As the wind fields at different height layers are different,
some individual observations received a negative transportation value,
which meant that the pollutants are transported from west of the route
to east. On the contrary, positive values indicate that the pollutants are
transported in the opposite direction. Fig. 12(b) shows the transporta-
tion of each individual observation segment and each height layer,
which are obtained by the model simulated NO2 VCDs and wind field
information. With more information on the vertical direction in the
(5) second method, it can be observed that the NO2 transportation is
mainly focused in the range from the surface to the height of 500 m,
and the highest transportation value is located at the height of about
200 m. These vertical structural characteristics of NO2 transportation
are significantly influenced by the vertical distribution characteristics
of NO2 and wind field.
The total transportation flux of this sectional segment is about
7.12 kg s−1 and 7.09 kg s−1 using the measured and simulated NO2
VCDs, respectively. The similar results of the two methods revealed that

10
W. Tan, et al.
Fig. 11. A sketch map of transportation estimation, the cruise track and NO2
VCDs (on 15 January 2018) of mobile MAX-DOAS measurements are marked by
colored line, while the wind field information is indicated by the black arrows.
the total NO2 transportation was reasonable. As shown in Fig. 13, we
compared the individual transportation of these two methods. Both the
distance-series plot and the correlation analysis (R = 0.89) show a good
agreement of these two different methods. This means that the east
wind brings about 7.12 kg NO2 per second from the right side to the left
side of this measurement route. The transportation flux of this magni-
tude is significant and considerable for affecting the ambient air en-
vironment, thereby suggesting that the quantification of regional
transportation and joint management are very important for the
Fig. 12. The transportations of individual observation segment, which (a) is calculated by mobile MAX-DOAS measured NO2 VCDs and model simulated wind
information, while (b) is calculated by model simulated NO2 VCD and wind information.
11
Atmospheric Research 245 (2020) 105037
prevention and control of regional air pollution.
3.4. Power plant NO2 and SO2 emissions
During the mobile observations, the observation vehicle passed by
some important emission sources, e.g., power plants and chemical in-
dustries. In particular, the power generation process of a thermal power
plant will have a significant impact on the ambient atmospheric en-
vironment, for instance, the pollution of particulate matter, sulfide,
nitrogen oxides, and others will be produced during the coal combus-
tion process (Lee et al., 1997; Frins et al., 2014). As the thermal power
plant is an important source for the pollution of the surroundings, it is
necessary to conduct quantitative analysis of power plant emissions.
Two typical power plants, namely the Dezhou Power Plant (116.25°
E, 37.44° N) and Hengshui Power Plant (115.60° E, 37.76° N) were
selected to estimate the NO2 and SO2 emissions (Table 3). For the
Dezhou Power Plant observations, we have chosen the observational
method to scan the cross section of the plume in the downwind area (Li
et al., 2007). As shown in Fig. 14(a) and (b), the vehicle passed through
the downwind area of the power plant during the observations on 27
January 2018, where the suddenly significant changes in NO2 and SO2
VCDs were observed. Accordingly, the boundary of the plume cross
section was obtained in Fig. 14(c) and (d), which is also verified by the
estimation of the plume using the Gaussian diffusion model. Based on
the calculation process described in Section 2.3, the NO2 and SO2
emission fluxes of the Dezhou Power Plant are 0.79 kg s−1 and
1.11 kg s−1, respectively. Moreover, another widely used observational
method of encircling the emission source was used for the measure-
ments at the Hengshui Power Plant (Wang et al., 2012; Wu et al., 2013;
Shaiganfar et al., 2017; Tan et al., 2019). As shown in Fig. 15, the
Hengshui Power Plant was encircled observation for twice on 9 Feb-
ruary 2018. Through integrating along the encircled route, the total
emissions of NO2 and SO2 can be obtained. During these two
W. Tan, et al. Atmospheric Research 245 (2020) 105037

Fig. 13. Comparison of the transportation results obtained by two different estimation methods, while (a) is the distance series plot and (b) is the correlation analysis
of the transportations obtained by these two different methods.

Table 3
The emission results of NO2 and SO2 from two different power plants.
References Observation Date Power plant Desulphurization Total installed capacity (MW) NO2 emissions (kg s−1) SO2 emissions (kg s−1)

This study 27 Jan. 2018 Dezhou, China Yes 2670 0.79 1.11
09 Feb. 2018 Hengshui, China Yes 1260 0.12 0.36
(Wu et al., 2017a) Oct. to Nov. 2011 Shijiazhuang, China Yes 600 0.107(±0.04) 0.078 (±0.036) (CEMS)a
(Frins et al., 2014) Mar. 2012 Montevideo, Uruguay No 385 11(±3) × 10−3 0.36 ± 0.12

a
Emission achieved from Continuous Emission Monitoring System (CEMS).

Fig. 14. The plume cross section scanning ob-

servations of the Dezhou power plant, (a) and (b)
are the NO2and SO2VCDs distributions near the
Dezhou power plant, while (c) and (d) are the
segments of the plume cross section for emission
fluxes calculation of NO2 and SO2, respectively.
The wind field information is marked by black
arrows and the power plant position is marked
by the red point.

observation periods, the averaged NO2 and SO2 emission fluxes of this
power plant are 0.12 kg s−1 and 0.36 kg s−1, respectively. The total
installed capacity of the Dezhou Power Plant is about 2.12 times higher
than that of the Hengshui Power Plant (Table 3), but the emission fluxes
of the Dezhou Power Plant are about 6.6 (NO2 emission fluxes) and 3.1
(SO2 emission fluxes) times higher than those of the Hengshui Power
Plant. Besides the difference in the total installed capacity, the different
observation methods also can introduce some deviations. The plume
cross section scanning method cannot distinguish between the pollu-
tants bring from upwind areas, power plant emissions, and even from
other sources if there are considerable pollutants surrounding the
emission source, while the encircled observational method can obtain
the net emission flux through subtracting the influx from the outflux.
Therefore, the emission flux of the Dezhou Power Plant may be slightly
overestimated owing to the plume cross section scanning observation.
Nevertheless, the results of the power plant emissions from the

12
W. Tan, et al.
mobile MAX-DOAS observations are reasonable and comparable to
those of previous mobile DOAS studies listed in Table 3; for example,
the NO2 emission flux of the Hengshui Power Plant is close to that of a
power plant of Shijiazhuang (Wu et al., 2017a). However, the total
installed capacity of the Hengshui Power Plant is about 2.1 times higher
than that of the power plant in Shijiazhuang. This means that the ef-
fective denitrification measures in these years have reduced the NO2
emissions by nearly 46.6% for the Hengshui Power Plant. However,
owing to the different research methods, we obtain a different SO2
emissions by the mobile MAX-DOAS observations in this study compare
to those obtained in the Continuous Emission Monitoring System
(CEMS) study of the power plant in Shijiazhuang. Moreover, the SO2
emissions of the Hengshui Power Plant are consistent with those of the
power plant in Montevideo, Uruguay (Frins et al., 2014). As the total
installed capacity of the Hengshui Power Plant is about 3.3 times higher
than that of the power plant in Montevideo, the effective desulfuriza-
tion measures of the Hengshui Power Plant have reduced the SO2
emissions by nearly 69.4%. Therefore, it is important to implement
denitrification and desulfurization processes before the emission of
exhaust from power plants. In recent years, the Chinese government
and relevant departments have formulated increasingly strict emission
reduction measures, and desulfurization and denitrification equipment
is almost universal. Power plant emissions have now reached or even
exceeded the international standards (Schreifels et al., 2012; Xue et al.,
2016)
4. Conclusions
Large-scale mobile MAX-DOAS observations were performed over
the NCP region in both summer and winter. Through the mobile ob-
servations, the distribution of NO2 and SO2 VCDs was obtained along
the measurement routes. In addition, we also quantitatively analyzed
the pollutant transportation and emissions of different sources.
Through the spectral fitting and AMF conversion process, spatio-
temporal distribution information of NO2 and SO2 VCDs was obtained
over this region along with the GPS position. The measured average
NO2 VCD is 35.97 × 1015 molec cm−2 in winter and 11.41 × 1015
molec cm−2 in summer, with a large variation between these two
seasons. In addition, the mean value of SO2 VCD is 29.47 × 1015 molec
13
Atmospheric Research 245 (2020) 105037
Fig. 15. The encircled observations of the Hengshui
power plant, (a) and (b) are the first time encircled
observation's NO2 and SO2 VCDs distributions near
the Hengshui power plant, while (c) and (d) are the
second circle. The wind field information is marked
by black arrows and the power plant position is
marked by the red point.
cm−2 during the winter observations. The comparison of the mobile
observations with the OMI tropospheric NO2 products show a good
agreement with a high Pearson correlation coefficient (R) of 0.92. In
addition, the comparison of in-situ monitoring data and mobile mea-
surements present a highly consistent variation trend, especially in the
comparison of NO2 concentrations. The ratios of SO2 to NO2 reveal the
emission characteristics of different cities.
High concentrations of both NO2 and SO2 were detected in northern
Anhui Province, central Shandong Province, and the Beijing-Tianjin-
Hebei region during the winter observations in January and February
2018. Combined with the WRF-Chem model simulated wind field in-
formation, the NO2 transportation from north Jiangsu Province to
Anhui Province is quantified as 7.12 kg s−1. This meant that about
7.12 kg NO2 from north Jiangsu Province is transported to north Anhui
Province under the action of easterly wind during the observation
period, thereby indicating the significant impacts of regional trans-
portation of ambient air pollutants during the large-scale pollution
event.
Finally, we estimated the NO2 and SO2 emissions of the Dezhou and
Hengshui power plants by scanning the plume cross section and en-
circling the emission source, respectively. The NO2 and SO2 emission
fluxes of the Dezhou Power Plant are 0.79 kg s−1 and 1.11 kg s−1,
respectively, while the NO2 and SO2 emission fluxes of the Hengshui
Power Plant are 0.12 kg s−1 and 0.36 kg s−1, respectively. These
emissions are comparable with those of previous power plant studies,
and the comparison also highlights the importance of implementing
desulfurization and denitrification processes before the emission of
exhaust from power plants.
Data availability
All data used in this study are available for scientific purposes upon
request to the corresponding authors (shanshanwang@fudan.edu.cn
and chliu81@ustc.edu.cn).
Author contributions
WT, CL, and SSW designed and implemented the research and wrote
the paper; WT, and HRL carried out the mobile MAX-DOAS
W. Tan, et al.
observations; YZZ conducted the WRF-chem model simulation; WJS
performed the satellite data retrieval; QHH and JGL contributed to
discuss the results.
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
Acknowledgements
This work was supported by grants from National Key Research and
Development Program of China (grant nos., 2018YFC0213104,
2018YFC0213201, 2016YFC0203302 and 2017YFC0210002), Anhui
Science and Technology Major Project (18030801111), the National
Natural Science Foundation of China (grant nos. 41722501, 91544212,
51778596, 41575021 and 41977184), the Strategic Priority Research
Program of the Chinese Academy of Sciences (grant nos.
XDA23020301), the National Key Project for Causes and Control of
Heavy Air Pollution (grant nos. DQGG0102-03 and DQGG0205), the
National High-Resolution Earth Observation Project of China (grant
nos. 05-Y30B01-9001-19/20-3) and the Civil Aerospace Technology
Advance Research Project, No. Y7K00100KJ. We acknowledge the
Ministry of Environment Protection of the People's Republic of China
for making data measured by the atmospheric environment automatic
monitoring stations available online. We thank the OMI International
Science Team for making OMI L1B data available for scientific re-
searches. We thank WRF-Chem developers for providing the model
available to the scientific community. We also thank the Belgian
Institute for Space Aeronomy (BIRA-IASB), Brussels, Belgium, for their
freely accessible QDOAS software.
References
Kreher, K., Van Roozendael, M., Hendrick, F., Apituley, A., Dimitropoulou, E., Frieß, U.,
Richter, A., Wagner, T., Abuhassan, N., Ang, L., Anguas, M., Bais, A., Benavent, N.,
Bösch, T., Bognar, K., Borovski, A., Bruchkouski, I., Cede, A., Chan, K.L., Donner, S.,
Drosoglou, T., Fayt, C., Finkenzeller, H., Garcia-Nieto, D., Gielen, C., Gómez-Martín,
L., Hao, N., Herman, J.R., Hermans, C., Hoque, S., Irie, H., Jin, J., Johnston, P.,
Khayyam Butt, J., Khokhar, F., Koenig, T.K., Kuhn, J., Kumar, V., Lampel, J., Liu, C.,
Ma, J., Merlaud, A., Mishra, A.K., Müller, M., Navarro-Comas, M., Ostendorf, M.,
Pazmino, A., Peters, E., Pinardi, G., Pinharanda, M., Piters, A., Platt, U., Postylyakov,
O., Prados-Roman, C., Puentedura, O., Querel, R., Saiz-Lopez, A., Schönhardt, A.,
Schreier, S.F., Seyler, A., Sinha, V., Spinei, E., Strong, K., Tack, F., Tian, X.,
Tiefengraber, M., Tirpitz, J.-L., van Gent, J., Volkamer, R., Vrekoussis, M., Wang, S.,
Wang, Z., Wenig, M., Wittrock, F., Xie, P.H., Xu, J., Yela, M., Zhang, C., Zhao, X.,
2019. Intercomparison of NO2, O4, O3 and HCHO slant column measurements by
MAX-DOAS and zenith-sky UV-Visible spectrometers during the CINDI-2 campaign.
Atmos. Meas. Tech. Discuss. https://doi.org/10.5194/amt-2019-157.
Pinardi, G., Van Roozendael, M., Abuhassan, N., Adams, C., Cede, A., Clémer, K., Fayt, C.,
Frieß, U., Gil, M., Herman, J., Hermans, C., Hendrick, F., Irie, H., Merlaud, A.,
Navarro Comas, M., Peters, E., Piters, A.J.M., Puentedura, O., Richter, A.,
Schönhardt, A., Shaiganfar, R., Spinei, E., Strong, K., Takashima, H., Vrekoussis, M.,
Wagner, T., Wittrock, F., Yilmaz, S., 2013. MAX-DOAS formaldehyde slant column
measurements during CINDI: intercomparison and analysis improvement. Atmos.
Meas. Tech. 6, 167–185. https://doi.org/10.5194/amt-6-167-2013.
Wang, Y., Dörner, S., Donner, S., Böhnke, S., De Smedt, I., Dickerson, R.R., Dong, Z., He,
H., Li, Z., Li, Z., Li, D., Liu, D., Ren, X., Theys, N., Wang, Y., Wang, Y., Wang, Z., Xu,
H., Xu, J., Wagner, T., 2019. Vertical profiles of NO2, SO2, HONO, HCHO, CHOCHO
and aerosols derived from MAX-DOAS measurements at a rural site in the central
western North China Plain and their relation to emission sources and effects of re-
gional transport. Atmos. Chem. Phys. 19, 5417–5449. https://doi.org/10.5194/acp-
19-5417-2019.
Brinksma, E.J., Pinardi, G., Volten, H., Braak, R., Richter, A., Schönhardt, A., Roozendael,
M.V., Fayt, C., Hermans, C., Dirksen, R.J., 2008. The 2005 and 2006 DANDELIONS
NO2 and aerosol intercomparison campaigns. J. Geophys. Res. 113, 762–770. https://
doi.org/10.1029/2007JD008808.
Bucsela, E.J., Krotkov, N.A., Celarier, E.A., Lamsal, L.N., 2013. A new stratospheric and
tropospheric NO2 retrieval algorithm for nadir-viewing satellite instruments: appli-
cations to OMI. Atmos. Meas. Tech. 6, 2607–2626. https://doi.org/10.5194/amt-6-
2607-2013.
Casaballe, N., Osorio, M., Di Martino, M., Frins, E., 2017. Comparison between
Regularized Optimization Algorithms for Tomographic Reconstruction of Plume
Cross Sections in the Atmosphere. Earth and Space Science 4, 723–736. https://doi.
14
Atmospheric Research 245 (2020) 105037
org/10.1002/2017ea000341.
Casaballe, N., Di Martino, M., Osorio, M., Ferrari, J.A., Wagner, T., Frins, E., 2020.
Improved algorithm with adaptive regularization for tomographic reconstruction of
gas distributions using DOAS measurements. ApOpt 59. https://doi.org/10.1364/ao.
383584.
Chan, K., 2017. Biomass burning sources and their contributions to the local air quality in
Hong Kong. Sci. Total Environ. 596, 212–221. https://doi.org/10.1016/j.scitotenv.
2017.04.091.
Chan, K.L., Hartl, A., Lam, Y.F., Xie, P.H., Liu, W.Q., Cheung, H.M., Lampel, J., Pöhler, D.,
Li, A., Xu, J., Zhou, H.J., Ning, Z., Wenig, M.O., 2015. Observations of tropospheric
NO2 using ground based MAX-DOAS and OMI measurements during the Shanghai
World Expo 2010. Atmos. Environ. 119, 45–58. https://doi.org/10.1016/j.atmosenv.
2015.08.041.
Chan, K.L., Wiegner, M., Wenig, M., Pöhler, D., 2017. Observations of tropospheric
aerosols and NO2 in Hong Kong over 5 years using ground based MAX-DOAS. Sci.
Total Environ. https://doi.org/10.1016/j.scitotenv.2017.10.153.
Chan, K.L., Wang, Z., Ding, A., Heue, K.-P., Shen, Y., Wang, J., Zhang, F., Shi, Y., Hao, N.,
Wenig, M., 2019. MAX-DOAS measurements of tropospheric NO2 and HCHO in
Nanjing and a comparison to ozone monitoring instrument observations. Atmos.
Chem. Phys. 19, 10051–10071. https://doi.org/10.5194/acp-19-10051-2019.
Chance, K., Kurucz, R.L., 2010. An improved high-resolution solar reference spectrum for
earth’s atmosphere measurements in the ultraviolet, visible, and near infrared. J.
Quant. Spectrosc. Ra. 111, 1289–1295. https://doi.org/10.1016/j.jqsrt.2010.01.036.
Chen, Z., Chen, D., Wen, W., Zhuang, Y., Kwan, M.-P., Chen, B., Zhao, B., Yang, L., Gao,
B., Li, R., Xu, B., 2019. Evaluating the “2+26” regional strategy for air quality im-
provement during two air pollution alerts in Beijing: variations in PM2.5 concentra-
tions, source apportionment, and the relative contribution of local emission and re-
gional transport. Atmos. Chem. Phys. 19, 6879–6891. https://doi.org/10.5194/acp-
19-6879-2019.
Clémer, K., Van Roozendael, M., Fayt, C., Hendrick, F., Hermans, C., Pinardi, G., Spurr, R.,
Wang, P., De Mazière, M., 2010. Multiple wavelength retrieval of tropospheric
aerosol optical properties from MAXDOAS measurements in Beijing. Atmos. Meas.
Tech. 3, 863–878. https://doi.org/10.5194/amt-3-863-2010.
Fleischmann, O.C., Hartmann, M., 2004. New ultraviolet absorption cross-sections of BrO
at atmospheric temperatures measured by time-windowing Fourier transform spec-
troscopy. J. Photoch. Photobio. A 168, 117–132.
Frins, E., Bobrowski, N., Osorio, M., Casaballe, N., Belsterli, G., Wagner, T., Platt, U.,
2014. Scanning and mobile multi-axis DOAS measurements of SO2 and NO2 emissions
from an electric power plant in Montevideo, Uruguay. Atmos. Environ. 98, 347–356.
https://doi.org/10.1016/j.atmosenv.2014.03.069.
Fu, G.Q., Xu, W.Y., Yang, R.F., Li, J.B., Zhao, C.S., 2014. The distribution and trends of fog
and haze in the North China Plain over the past 30 years. Atmos. Chem. Phys. 14,
11949–11958. https://doi.org/10.5194/acp-14-11949-2014.
Gao, M., Carmichael, G.R., Wang, Y., Saide, P.E., Yu, M., Xin, J., Liu, Z., Wang, Z., 2016.
Modeling study of the 2010 regional haze event in the North China Plain. Atmos.
Chem. Phys. 16, 1673–1691. https://doi.org/10.5194/acp-16-1673-2016.
Grell, G.A., Peckhama, S.E., Schmitzc, R., McKeenb, S.A., Frostb, G., William, C.S., Eder,
B., 2005. Fully coupled “online” chemistry within the WRF model. Atmos. Environ.
39 (37), 6957–6975. https://doi.org/10.1016/j.atmosenv.2005.04.027.
Halla, J.D., Wagner, T., Beirle, S., Brook, J.R., Hayden, K.L., Brien, J.M., Ng, A., Majonis,
D., Wenig, M.O., McLaren, R., 2011. Determination of tropospheric vertical columns
of NO2 and aerosol optical properties in a rural setting using MAX-DOAS. Atmos.
Chem. Phys. 11, 12475–12498. https://doi.org/10.5194/acp-11-12475-2011.
Hendrick, F., Müller, J.F., Clémer, K., Wang, P., De Mazière, M., Fayt, C., Gielen, C.,
Hermans, C., Ma, J.Z., Pinardi, G., Stavrakou, T., Vlemmix, T., Van Roozendael, M.,
2014. Four years of ground-based MAX-DOAS observations of HONO and NO2 in the
Beijing area. Atmos. Chem. Phys. 14, 765–781. https://doi.org/10.5194/acp-14-765-
2014.
Hong, Q., Liu, C., Chan, K.L., Hu, Q., Xie, Z., Liu, H., Si, F., Liu, J., 2018. Ship-based MAX-
DOAS measurements of tropospheric NO2, SO2, and HCHO distribution along the
Yangtze River. Atmos. Chem. Phys. 18, 5931–5951. https://doi.org/10.5194/acp-18-
5931-2018.
Hönninger, G., von Friedeburg, C., Platt, U., 2004. Multi axis differential optical ab-
sorption spectroscopy (MAX-DOAS). Atmos. Chem. Phys. 4, 231–254. https://doi.
org/10.5194/acp-4-231-2004.
Huang, R.-J., Zhang, Y., Bozzetti, C., Ho, K.-F., Cao, J.-J., Han, Y., Daellenbach, K.R.,
Slowik, J.G., Platt, S.M., Canonaco, F., 2014. High secondary aerosol contribution to
particulate pollution during haze events in China. Nature 514, 218. https://doi.org/
10.1038/nature13774.
Ibrahim, O., Shaiganfar, R., Sinreich, R., Stein, T., Platt, U., Wagner, T., 2010. Car MAX-
DOAS measurements around entire cities: quantification of NOx emissions from the
cities of Mannheim and Ludwigshafen (Germany). Atmos. Meas. Tech. 3, 709–721.
https://doi.org/10.5194/amt-3-709-2010.
Irie, H., Takashima, H., Kanaya, Y., Boersma, K.F., Gast, L., Wittrock, F., Brunner, D.,
Zhou, Y., Van Roozendael, M., 2011. Eight-component retrievals from ground-based
MAX-DOAS observations. Atmos. Meas. Tech. 4, 1027–1044. https://doi.org/10.
5194/amt-4-1027-2011.
Jang, M., Kamens, R.M., 2001. Characterization of secondary aerosol from the photo-
oxidation of toluene in the presence of NOx and 1-propene. Environ. Sci. Technol. 35,
3626–3639. https://doi.org/10.1021/es010676+.
Jin, J., Ma, J., Lin, W., Zhao, H., Shaiganfar, R., Beirle, S., Wagner, T., 2016. MAX-DOAS
measurements and satellite validation of tropospheric NO2 and SO2 vertical column
densities at a rural site of North China. Atmos. Environ. 133, 12–25. https://doi.org/
10.1016/j.atmosenv.2016.03.031.
Krotkov, N.A., Lamsal, L.N., Celarier, E.A., Swartz, W.H., Marchenko, S.V., Bucsela, E.J.,
Chan, K.L., Wenig, M., Zara, M., 2017. The version 3 OMI NO2 standard product.
W. Tan, et al.
Atmos. Meas. Tech. 10, 3133–3149. https://doi.org/10.5194/amt-10-3133-2017.
Lee, D.S., Köhler, I., Grobler, E., Rohrer, F., Sausen, R., Gallardo-Klenner, L., Olivier,
J.G.J., Dentener, F.J., Bouwman, A.F., 1997. Estimations of global no, emissions and
their uncertainties. Atmos. Environ. 31, 1735–1749.
Levelt, P.F., Oord, G.H.J.V., Dobber, M.R., Malkki, A., Visser, H., Vries, J.D., Stammes, P.,
Lundell, J.O.V., Saari, H., 2006. The ozone monitoring instrument. IEEE T. Geosci.
Remote Sens 44, 1093–1101. https://doi.org/10.1109/TGRS.2006.872333.
Li, A., Xie, P., Liu, W., 2007. Monitoring of total emission volume from pollution sources
based on passive differential optical absorption spectroscopy. Acta Optica Sinic 27,
1537.
Liu, H., Cheng, L., Xie, Z., Ying, L., Xin, H., Wang, S., Jin, X., Xie, P., 2016. A paradox for
air pollution controlling in China revealed by “APEC Blue” and “Parade Blue” Sci.
Rep. 6, 34408. https://doi.org/10.1038/srep34408.
Ma, J.Z., Beirle, S., Jin, J.L., Shaiganfar, R., Yan, P., Wagner, T., 2013. Tropospheric NO2
vertical column densities over Beijing: results of the first three years of ground-based
MAX-DOAS measurements (2008-2011) and satellite validation. Atmos. Chem. Phys.
13, 1547–1567. https://doi.org/10.5194/acp-13-1547-2013.
Meller, R., Moortgat, G.K., 2000. Temperature dependence of the absorption cross sec-
tions of formaldehyde between 223 and 323 K in the wavelength range 225–375 nm.
J. Geophys. Res.-Atmos. 105, 7089–7101.
MEP: 2016, 2017. Report on the State of the Environment in China, MEP, Beijing, China.
available at: http://english.mee.gov.cn/Resources/Reports/soe/ReportSOE/201709/
P020170929573904364594.pdf, Accessed date: 13 April 2019 (in Chinese).
Ortega, I., Koenig, T., Sinreich, R., Thomson, D., Volkamer, R., 2015. The CU 2-D-MAX-
DOAS instrument – part 1: Retrieval of 3-D distributions of NO2 and azimuth-de-
pendent OVOC ratios. Atmos. Meas. Tech. 8, 2371–2395. https://doi.org/10.5194/
amt-8-2371-2015.
Platt, U., Stutz, J., 2008. Differential Optical Absorption Spectroscopy. Springer, Berlin
Heidelberg, pp. 2458–2462.
Schreifels, J.J., Fu, Y., Wilson, E.J., 2012. Sulfur dioxide control in China: policy evolu-
tion during the 10th and 11th Five-year Plans and lessons for the future. Energy
Policy 48, 779–789. https://doi.org/10.1016/j.enpol.2012.06.015.
Serdyuchenko, A., Gorshelev, V., Weber, M., Chehade, W., Burrows, J.P., 2014. High
spectral resolution ozone absorption cross-sections - part 2: Temperature dependence.
Atmos. Meas. Tech. 7, 625–636. https://doi.org/10.5194/amt-7-625-2014.
Seyler, A., Meier, A.C., Wittrock, F., Kattner, L., Mathieu-Üffing, B., Peters, E., Richter, A.,
Ruhtz, T., Schönhardt, A., Schmolke, S., Burrows, J.P., 2019. Studies of the horizontal
inhomogeneities in NO2 concentrations above a shipping lane using ground-based
multi-axis differential optical absorption spectroscopy (MAX-DOAS) measurements
and validation with airborne imaging DOAS measurements. Atmos. Meas. Tech. 12,
5959–5977. https://doi.org/10.5194/amt-12-5959-2019.
Shaiganfar, R., Beirle, S., Denier Van Der Gon, H., Jonkers, S., Kuenen, J., Petetin, H.,
Zhang, Q., Beekmann, M., Wagner, T., 2017. Estimation of the Paris NOx emissions
from mobile MAX-DOAS observations and CHIMERE model simulations during the
MEGAPOLI campaign using the closed integral method. Atmos. Chem. Phys. 17,
7853–7890. https://doi.org/10.5194/acp-17-7853-2017.
Shaiganfar, R., Beirle, S., Sharma, M., Chauhan, A., Singh, R.P., Wagner, T., 2011.
Estimation of NOx emissions from Delhi using Car MAX-DOAS observations and
comparison with OMI satellite data. Atmos. Chem. Phys. 11, 10871–10887. https://
doi.org/10.5194/acp-11-10871-2011.
Smekens, J.-F., Burton, M.R., Clarke, A.B., 2015. Validation of the SO2 camera for high
temporal and spatial resolution monitoring of SO2 emissions. JVGR 300, 37–47.
https://doi.org/10.1016/j.jvolgeores.2014.10.014.
Su, J., 2018. Portable and sensitive air pollution monitoring. Light: Science &
Applications 7. https://doi.org/10.1038/s41377-018-0017-x.
Su, W., Liu, C., Hu, Q., Fan, G., Xie, Z., Huang, X., Zhang, T., Chen, Z., Dong, Y., Ji, X., Liu,
H., Wang, Z., Liu, J., 2017. Characterization of ozone in the lower troposphere during
the 2016 G20 conference in Hangzhou. Sci. Rep. 7, 17368. https://doi.org/10.1038/
s41598-017-17646-x.
Sun, Y., Qi, J., Wang, Z., Fu, P., Jie, L., Yang, T., Yan, Y., 2014. Investigation of the
sources and Evolution Processes of Severe Haze Pollution in Beijing in January 2013.
J. Geophys. Res.-Atmos. 119, 4380–4398. https://doi.org/10.1002/2014JD021641.
Tan, W., Liu, C., Wang, S., Xing, C., Su, W., Zhang, C., Xia, C., Liu, H., Cai, Z., Liu, J.,
2018. Tropospheric NO2, SO2, and HCHO over the East China Sea, using ship-based
MAX-DOAS observations and comparison with OMI and OMPS satellites data. Atmos.
Chem. Phys. 18, 15387–15402. https://doi.org/10.5194/acp-18-15387-2018.
Tan, W., Zhao, S., Liu, C., Chan, K.L., Xie, Z., Zhu, Y., Su, W., Zhang, C., Liu, H., Xing, C.,
Liu, J., 2019. Estimation of winter time NOx emissions in Hefei, a typical inland city
of China, using mobile MAX-DOAS observations. Atmos. Environ. 200, 228–242.
https://doi.org/10.1016/j.atmosenv.2018.12.009.
Tao, M., Chen, L., Su, L., Tao, J., 2012. Satellite observation of regional haze pollution
over the North China Plain. J. Geophys. Res.-Atmos 117, D12203. https://doi.org/10.
15
Atmospheric Research 245 (2020) 105037
1029/2012jd017915.
Thalman, R., Volkamer, R., 2013. Temperature dependent absorption cross-sections of
O2-O2 collision pairs between 340 and 630 nm and at atmospherically relevant
pressure. Phys. Chem. Chem. Phys. 15, 15371–15381. https://doi.org/10.1039/
C3CP50968K.
Theys, N., De Smedt, I., van Gent, J., Danckaert, T., Wang, T., Hendrick, F., Stavrakou, T.,
Bauduin, S., Clarisse, L., Li, C., Krotkov, N., Yu, H., Brenot, H., Van Roozendael, M.,
2015. Sulfur dioxide vertical column DOAS retrievals from the ozone monitoring
Instrument: Global observations and comparison to ground-based and satellite data.
J. Geophys. Res.-Atmos. 120, 2470–2491. https://doi.org/10.1002/2014jd022657.
Vandaele, A.C., Hermans, C., Simon, P.C., Carleer, M., Colin, R., Fally, S., Mérienne, M.F.,
Jenouvrier, A., Coquart, B., 1998. Measurements of the NO2 absorption cross-section
from 42000 cm−1 to 10000 cm−1 (238–1000 nm) at 220 K and 294 K. J. Quant.
Spectrosc. Ra. 59, 171–184.
Vandaele, A.C., Hermans, C., Fally, S., 2009. Fourier transform measurements of SO2
absorption cross sections: II. : Temperature dependence in the 29000–44000 cm−1
(227–345 nm) region. J. Quant. Spectrosc. Ra. 110, 2115–2126.
Wagner, T., Ibrahim, O., Shaiganfar, R., Platt, U., 2010. Mobile MAX-DOAS observations
of tropospheric trace gases. Atmos. Meas. Tech. 2, 129–140. https://doi.org/10.
5194/amt-3-129-2010.
Wang, L., Xu, J., Yang, J., Zhao, X., Wei, W., Cheng, D., Pan, X., Su, J., 2012.
Understanding haze pollution over the southern Hebei area of China using the CMAQ
model. Atmos. Environ. 56, 69–79. https://doi.org/10.1016/j.atmosenv.2012.04.
013.
Wang, T., Hendrick, F., Wang, P., Tang, G., Clémer, K., Yu, H., Fayt, C., Hermans, C.,
Gielen, C., Müller, J.F., Pinardi, G., Theys, N., Brenot, H., Van Roozendael, M., 2014.
Evaluation of tropospheric SO2 retrieved from MAX-DOAS measurements in Xianghe,
China. Atmos. Chem. Phys. 14, 11149–11164. https://doi.org/10.5194/acp-14-
11149-2014.
Wang, P., Piters, A., van Geffen, J., Tuinder, O., Stammes, P., Kinne, S., 2020. Shipborne
MAX-DOAS measurements for validation of TROPOMI NO2 products. Atmos. Meas.
Tech. 13, 1413–1426. https://doi.org/10.5194/amt-13-1413-2020.
Wu, F., Li, A., Xie, P., Chen, H., Hu, Z., Zhang, Q., Liu, J., Liu, W., 2017a. Emission flux
measurement error with a mobile DOAS system and application to NOx flux ob-
servations. Sensors (Basel) 17. https://doi.org/10.3390/s17020231.
Wu, F., Xie, P., Li, A., Mou, F., Chen, H., Zhu, Y., Zhu, T., Liu, J., Liu, W., 2018.
Investigations of temporal and spatial distribution of precursors SO2 and NO2 vertical
columns in the North China Plain using mobile DOAS. Atmos. Chem. Phys. 18,
1535–1554. https://doi.org/10.5194/acp-18-1535-2018.
Xing, J., Wang, S.X., Chatani, S., Zhang, C.Y., Wei, W., Hao, J.M., Klimont, Z., Cofala, J.,
Amann, M., 2011. Projections of air pollutant emissions and its impacts on regional
air quality in China in 2020. Atmos. Chem. Phys. 11, 3119–3136. https://doi.org/10.
5194/acp-11-3119-2011.
Xing, C., Liu, C., Wang, S., Chan, K.L., Gao, Y., Huang, X., Su, W., Zhang, C., Dong, Y.,
Fan, G., Zhang, T., Chen, Z., Hu, Q., Su, H., Xie, Z., Liu, J., 2017. Observations of the
vertical distributions of summertime atmospheric pollutants and the corresponding
ozone production in Shanghai, China. Atmos. Chem. Phys. 17, 14275–14289. https://
doi.org/10.5194/acp-17-14275-2017.
Xu, W.Y., Zhao, C.S., Ran, L., Deng, Z.Z., Liu, P.F., Ma, N., Lin, W.L., Xu, X.B., Yan, P., He,
X., Yu, J., Liang, W.D., Chen, L.L., 2011. Characteristics of pollutants and their cor-
relation to meteorological conditions at a suburban site in the North China Plain.
Atmos. Chem. Phys. 11, 4353–4369. https://doi.org/10.5194/acp-11-4353-2011.
Xue, Y., Tian, H., Yan, J., Zhou, Z., Wang, J., Nie, L., Pan, T., Zhou, J., Hua, S., Wang, Y.,
Wu, X., 2016. Temporal trends and spatial variation characteristics of primary air
pollutants emissions from coal-fired industrial boilers in Beijing, China. Environ.
Pollut. 213, 717–726. https://doi.org/10.1016/j.envpol.2016.03.047.
Zhang, J., Van Der, A., Ding, J., 2018. Detection and emission estimates of NOx sources
over China North Plain using OMI observations. Int. J. Remote Sens. 39, 2847–2859.
https://doi.org/10.1080/01431161.2018.1430402.
Zhang, C., Liu, C., Hu, Q., Cai, Z., Su, W., Xia, C., Zhu, Y., Wang, S., Liu, J., 2019. Satellite
UV-Vis spectroscopy: implications for air quality trends and their driving forces in
China during 2005–2017. Light: Science & Applications 8, 1–12. https://doi.org/10.
1038/s41377-019-0210-6.
Zhao, X.J., Zhao, P.S., Xu, J., Meng, W., Pu, W.W., Dong, F., He, D., Shi, Q.F., 2013.
Analysis of a winter regional haze event and its formation mechanism in the North
China Plain. Atmos. Chem. Phys. 13, 5685–5696. https://doi.org/10.5194/acp-13-
5685-2013.
Zhou, D., Ding, K., Huang, X., Liu, L., Liu, Q., Xu, Z., Jiang, F., Fu, C., Ding, A., 2018.
Transport, mixing and feedback of dust, biomass burning and anthropogenic pollu-
tants in eastern Asia: a case study. Atmos. Chem. Phys. 18, 16345–16361. https://doi.
org/10.5194/acp-18-16345-2018.